This is an open access article published under an ACS AuthorChoice License, which permits copying and redistribution of the article or any adaptations for non-commercial purposes.
Article Cite This: ACS Omega 2017, 2, 8308-8312
http://pubs.acs.org/journal/acsodf
Circular Dichroistic Impacts of 1‑(3-Dimethylaminopropyl)-3ethylurea: Secondary Structure Artifacts Arising from Bioconjugation Using 1‑Ethyl-3-[3-dimethylaminopropyl]carbodiimide Matthew B. Kubilius and Raymond S. Tu* Chemical Engineering Department, City College of New York, 160 Convent Avenue, New York, New York 10031, United States S Supporting Information *
ABSTRACT: 1-Ethyl-3-[3-dimethylaminopropyl]carbodiimide (EDC) is a commonly used reagent for bioconjugation and peptide synthesis. Both EDC and the corresponding urea derivative, 1-(3-dimethylaminopropyl)-3-ethylurea (EDU), are achiral. As the reagent is active in aqueous solutions, it is a common choice for the study of evolving secondary structural changes via circular dichroism. This work highlights the effect of EDU on spectropolarimetric measurements, namely, the problematic absorption profile at low wavelengths (190−220 nm). We demonstrate that EDU is capable of erroneously indicating structural changes, particularly loss of α-helical character, through masking of the characteristic minimum at 208 nm. However, if the concentrations of the EDU in the sample are known, then this effect can be anticipated and calculations of secondary structure can be adjusted to avoid the impacted wavelengths. Impacts of EDU in a sample are compared to those of standard urea, which, by contrast, is commonly used as a denaturant in circular dichroism studies without issue.
■
INTRODUCTION This work highlights the influence of 1-(3-dimethylaminopropyl)-3-ethylurea (EDU) in circular dichroism (CD) measurements. The importance of this work stems from the ubiquity of the carbodiimide coupling agent, 1-ethyl-3-[3dimethylaminopropyl]carbodiimide (EDC) in bioconjugation.1−5 The presence of EDU in CD samples can cause the underlying data to be obscured, preventing the interpretation of structure for protein and peptide bioconjugates. A key advantage of EDC bioconjugation is its compatibility with aqueous solutions, making its conjugation chemistry an attractive choice for study of evolving secondary structural changes using circular dichroism.6,7 For example, recent work has demonstrated a desire to study the self-assemblies of small (three amino acid) peptide sequences computationally and experimentally.8 With an understanding of the influence of the linking reagents on these measurements, these self-assemblies have the potential to be measured as they are grown. EDC is used as a linking reagent for condensation reactions linking amines to carboxylic acid groups, typically in aqueous media. The EDU product of this reaction is the urea form of the original carbodiimide molecule, as shown in Figure 1. Because of the molecule’s ability to react with both water and the molecule of interest, EDC usage protocols generally recommend a 10:1 molar ratio of EDC to each carboxylic acid, yielding these same elevated concentrations of EDU in solution post-reaction.9 Both EDC and EDU are achiral, and they might be assumed not to directly impact CD measurement. Unlike urea, however, EDU’s applicable concentration range, absorption impact, and wavelength range of interference on circular dichroism measure© 2017 American Chemical Society
ments are not commonly known. Establishing those parameters will allow for the development of well-controlled studies with clearly understood results using this reagent in solution. Typically, the issues explored in this work are eliminated in advance of structural studies by chromatography purification of samples post-conjugation. Although that method is undeniably effective, it is incompatible with kinetic studies or attempts to directly measure the change in the secondary structure as a function of reaction completion in real time. We characterize the effect of EDU in a solution by direct inspection of CD signals of the common reference samples bovine serum albumin (BSA) and pantolactone measured in increasing concentrations of both urea and EDU. By evaluating the two-component system, we quantify the degree that standard urea can be used without issue, whereas the EDU can be problematic over similar concentration ranges. Because circular dichroism measures the difference in absorbance of left- and right-polarized light,10−13 any components in a solution able to change this ratio will give rise to changes in the spectra obtained. In this work, we highlight the degree that EDU absorbs over a much broader range of wavelengths as well as the impact that this adsorption has on secondary structure interpretation. Because only chiral molecules or molecules with certain fixed orientations give rise to such a difference, small, achiral molecules do not generally impact circular dichroism measurements. However, if an achiral species is present in a sample that significantly absorbs Received: September 1, 2017 Accepted: October 13, 2017 Published: November 21, 2017 8308
DOI: 10.1021/acsomega.7b01288 ACS Omega 2017, 2, 8308−8312
ACS Omega
Article
Figure 1. Reaction of EDC to EDU, either through use as a linking reagent, or via direct reaction with the aqueous solution.
Figure 2. CD spectra of BSA at various concentrations of 1-(3-dimethylaminopropyl)-3-ethylurea.
Figure 3. CD spectra of BSA at various concentrations of urea.
present, the protein shows a clear α-helical character. But at EDU concentrations as low as 0.05 M, the characteristic minimum expected near 208 nm is completely obscured. As EDU, being achiral, has a zero-magnitude CD signal of its own (Supporting Information), this effect cannot be thought of as the summation of the CD spectra of the two molecules. At 0.1 M EDU, the minimum at 222 has shifted to a higher wavelength and noticeably decreased in intensity. This shift continues with increasing concentration until a sample concentration of 0.4 M EDU, at which point it is no longer possible to observe the helical signal associated with BSA. Note that the signal decays to zero
unpolarized light at the wavelength of the CD scan, it will absorb a greater fraction of the transmitted light, simply due to the greater availability of the light less absorbed by the chiral molecule. This absorption results in a reduction in the magnitude of the reported circular dichroism measurements because of the disproportionate absolute absorption.
■
RESULTS AND DISCUSSION To quantify the nature of EDU’s effect on a system, a series of circular dichroism spectra are obtained on bovine serum albumin (BSA) in deionized (DI) water (Figure 2). With no EDU 8309
DOI: 10.1021/acsomega.7b01288 ACS Omega 2017, 2, 8308−8312
ACS Omega
Article
Figure 4. 1-(3-Dimethylaminopropyl)-3-ethylurea interferes with pantolactone circular dichroism spectra as a function of EDU concentration. Inset: CD spectra of pantolactone at various concentrations of urea.
with increasing EDU concentration. As we see signal loss, rather than signal transition to a random coil configuration, we establish that the effect is due to EDU’s impact on the CD signal, as opposed to EDU denaturing the secondary structure of the protein molecules. To better understand the magnitude of this phenomenon, we perform the same experiment using urea (Figure 3). A small loss is seen around the 210 nm wavelength, but the signal persists. Urea is a known denaturant, and the purpose of Figure 3 is to highlight the relative intensity of the data in Figure 2. Given that denaturation is potentially a factor in the EDU results, we performed a second experiment in which the signal loss could be decoupled from secondary structure changes. We duplicated the CD experiments with (S)-(+)-pantolactone replacing the BSA (Figure 4). (S)-(+)-Pantolactone is a chiral molecule incapable of secondary structure-based signal loss,14 and it is often used as a circular dichroism standard with a known minimum at 219 nm. Neither urea nor EDU will have a structural effect on this molecule, and their effects on absorbance and CD signal can be decoupled from losses in the secondary structure. When combining (S)-(+)-pantolactone with EDU (Figure 4), a clear trend of complete data obfuscation is seen, analogous to that seen in Figure 2. From this, we conclude that EDU’s absorbance gives rise to these results, as opposed to structural changes arising from the EDU acting as a denaturant. The two-component system with pantolactone and urea corroborates the hypothesis (Figure 4, inset). At 0.4 M urea, only minor changes in the data are seen, reinforcing the notion that urea does not interfere with circular dichroism measurements at these wavelengths and concentrations. The effects of urea are seen in the wavelengths below 200 nm, which is consistent with the standard practice of anticipating urea effects in data below 210 nm. From this, we can infer that the subtle changes seen in Figure 3 centered on 210 nm are a result of urea-induced denaturation of α-helical structures into random coils. Figure 5 quantifies the mechanism behind these results, plotting the absorbance data of 0.1 M EDU, 0.1 M urea, and 1 mg/mL bovine serum albumin (BSA) samples from 200 to 250 nm in a 1 cm quartz cuvette. Concentrations and units were
Figure 5. Absorbance profiles of 0.1 M EDU (red circles), 0.1 M urea (blue triangles), and 1 mg/mL bovine serum albumin (black squares) measured in a 1 cm quartz cuvette.
chosen to match those typically studied in circular dichroism experiments. Urea is seen with its characteristic high absorbance at low wavelengths, preventing reliable data collection below 210 nm. EDU, by contrast, absorbs heavily over the entire range of interest. Although this molecule is capable of impacting readings over the entire range, note that the error would be gradually, increasingly prominent at lower wavelengths. This results in the EDU yielding circular dichroism errors even at concentrations well below those deemed acceptable for urea. From these results, we suggest a new “rule of thumb” for the use of EDC in bioconjugation when paired with circular dichroism. At low (
